Hypotonically Induced Calcium Release from Intracellular Calcium Stores*

Osmotic cell swelling induced by hypotonic stress is associated with a rise in intracellular Ca 2 (cid:49) concentration, which is at least partly due to a release of Ca 2 (cid:49) from internal stores. Since osmotic influx of water di- lutes the cytoplasmic milieu, we have investigated how nonmitochondrial Ca 2 (cid:49) stores in permeabilized A7r5 cells respond to a reduction in cytoplasmic tonicity. We now present experimental evidence for a direct Ca 2 (cid:49) release from the stores when exposed to a hypotonic medium. The release is graded, but does not occur through the inositol trisphosphate or the ryanodine re- ceptor. Ca 2 (cid:49) seems to be released through the passive leak pathway, and this phenomenon can be partially inhibited by divalent cations in the following order of potency: Ni 2 (cid:49) Co 2 (cid:49) Mn 2 (cid:49) (cid:62) Mg 2 (cid:49) (cid:62) Ba 2 (cid:49) . This release also occurs in intact A7r5 cells. This novel mecha- nism of hypotonically induced Ca 2 (cid:49) release is therefore an inherent property of the stores, which can occur in the absence of second messengers. Intracellular stores can therefore act as osmosensors.


Osmotic cell swelling induced by hypotonic stress is associated with a rise in intracellular Ca
concentration, which is at least partly due to a release of Ca 2؉ from internal stores. Since osmotic influx of water dilutes the cytoplasmic milieu, we have investigated how nonmitochondrial Ca 2؉ stores in permeabilized A7r5 cells respond to a reduction in cytoplasmic tonicity. We now present experimental evidence for a direct Ca 2؉ release from the stores when exposed to a hypotonic medium. The release is graded, but does not occur through the inositol trisphosphate or the ryanodine receptor. Ca 2؉ seems to be released through the passive leak pathway, and this phenomenon can be partially inhibited by divalent cations in the following order of potency: Ni 2؉ ‫؍‬ Co 2؉ > Mn 2؉ > Mg 2؉ > Ba 2؉ . This release also occurs in intact A7r5 cells. This novel mechanism of hypotonically induced Ca 2؉ release is therefore an inherent property of the stores, which can occur in the absence of second messengers. Intracellular stores can therefore act as osmosensors. Most cells exposed to anisosmotic solutions activate volume regulatory processes to prevent damage by cell swelling or shrinkage (1)(2)(3). Osmotic cell swelling in response to hypotonic stress is associated with a rise in intracellular [Ca 2ϩ ], which is at least partly due to a release of Ca 2ϩ from internal stores (2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13). The link between changes in external osmotic concentration and internal Ca 2ϩ release is not known: inositol trisphosphate (3) and arachidonic acid (5) may be involved, but it is also possible that no messenger is needed (2). Since the influx of water dilutes the cytoplasm, we have investigated how the Ca 2ϩ stores in permeabilized A7r5 cells respond to a reduction in cytoplasmic tonicity. We now present experimental evidence for a direct Ca 2ϩ release from the nonmitochondrial stores when exposed to a hypotonic medium. This hypotonically induced Ca 2ϩ release through the passive leak pathway is an inherent property of the stores, which can occur in the absence of second messengers.

MATERIALS AND METHODS
A7r5 cells, an established cell line derived from embryonic rat aorta, were used between the 7th and the 18th passage after receipt from the American Type Culture Collection (Bethesda, MD) and subcultured weekly by trypsinization. The cells were cultured at 37°C in a 9% CO 2 incubator in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum, 3.8 mM L-glutamine, 0.9% (v/v) nonessential amino acids, 85 IU ml Ϫ1 penicillin, and 85 g ml Ϫ1 streptomycin. The cells were seeded for the 45 Ca 2ϩ fluxes in 12-well dishes (Costar, 4 cm 2 ) at a density of approximately 10 4 cells cm Ϫ2 and for the intracellular [Ca 2ϩ ] measurements in Coverglass Chambers (Nunc Inc., Naperville, IL) at a density of 5ϫ10 4 cells cm Ϫ2 . 45 Ca 2ϩ fluxes on monolayers of saponin-permeabilized A7r5 cells (3 ϫ 10 5 cells/4-cm 2 well) at 25°C were done as described (14). The stores were loaded for 40 min in 120 mM KCl, 30 mM imidazole (pH 6.8), 5 mM MgCl 2 , 5 mM ATP, 0.44 mM EGTA, 10 mM NaN 3 , and 150 nM free Ca 2ϩ (50 Ci ml Ϫ1 ). The wells were then washed twice in an isoosmotic efflux medium containing 60 mM KCl, 120 mM mannitol, 30 mM imidazole (pH 6.8), 1 mM EGTA, and 2 M thapsigargin (measured osmolality of 300 mosm/kg H 2 O). 1 ml of medium was then added at time 0 and replaced every 6 s or every 2 min. Osmolality changes were induced by changing the mannitol concentration to prevent changes in ionic concentration and ionic strength.
Single-cell intracellular [Ca 2ϩ ] measurements were performed using a laser-scanning MRC-1000 system (Bio-Rad, Hertfordshire, UK) attached to an inverted Nikon Diaphot 300 epifluorescence microscope with a CF Fluor 40 ϫ (numeric aperture ϭ 1.3) oil immersion objective. The cells were incubated for 30 min with 5 M Indo-1/AM dissolved in the modified Krebs solution and then further incubated for 1 to 2 h in the absence of Indo-1. During the experiment at 25°C, the cells were continuously superfused from a pipette placed on top of the cell. The solutions were the same as for the 45 Ca 2ϩ fluxes in intact cells.

RESULTS AND DISCUSSION
Permeabilized A7r5 cells loaded to equilibrium with 45 Ca 2ϩ slowly lost their 45 Ca 2ϩ during incubation in an isoosmotic (300 mosm/kg H 2 O) Ca 2ϩ -free medium. A long-lasting reduction in medium osmolality to 180 mosm/kg H 2 O transiently increased the rate of Ca 2ϩ release (Fig. 1A). This effect occurred despite the reduced concentration gradient for Ca 2ϩ across the store membrane as a result of the decreased luminal [Ca 2ϩ ] by osmotic H 2 O influx. This 40% reduction in tonicity is of the same order as used experimentally in intact cells (range 30 -50%, Refs. 4 -12). The release could be elicited again after reloading the stores with 45 Ca 2ϩ (data not shown). Increasing the osmolality to 420 mosm/kg H 2 O had no effect but the returning to isoosmotic solution again resulted in a transient, although less pronounced, Ca 2ϩ release (Fig. 1B).
Ca 2ϩ release through the inositol trisphosphate (InsP 3 ) 1 receptor (15) and ryanodine receptor (16,17) is graded, i.e. continuous submaximal stimulation is unable to completely empty the entire Ca 2ϩ pool. The hypotonically induced Ca 2ϩ release is also graded: a long-lasting moderate decrease in tonicity released less Ca 2ϩ than a more pronounced decrease ( Fig. 2A). * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
‡ To whom correspondence should be addressed. Graded responses allow increment detection (18). Increment detection for the hypotonically induced Ca 2ϩ release is shown in Fig. 2B: decreasing the osmolality stepwise resulted in concomitant phases of Ca 2ϩ release. The hypotonically induced Ca 2ϩ release (up to 60% of the total store Ca 2ϩ content in Fig. 2) originated from the InsP 3sensitive store, since the InsP 3 -insensitive compartment in permeabilized A7r5 cells contains only about 5% of the stored Ca 2ϩ (14). Activation of the InsP 3 receptor by endogenous InsP 3 formation was, however, not involved. First, a hypotonic challenge after 20 min (less filled stores) released more Ca 2ϩ and therefore resulted in a lower Ca 2ϩ content at 30 min than a challenge at 0 min (full stores, Fig. 3A). This larger Ca 2ϩ release from less filled stores is in contrast with the less complete InsP 3 -induced Ca 2ϩ release from partially depleted stores (14, 19 -21). A second argument against the involvement of the InsP 3 receptor is that 10 M thimerosal, a pharmacological activator of InsP 3 receptors (22-26), did not stimulate the hypotonically induced Ca 2ϩ release (Fig. 3B). A third argument indicating that the InsP 3 receptor was not involved is that the release was unaffected by the presence of 50 g ml Ϫ1 heparin (Fig. 3B, inset). Heparin is a competitive inhibitor of the InsP 3 receptor, which was found in control experiments to completely inhibit the Ca 2ϩ release in response to up to 3.  Ryanodine receptors were not involved in the release because 10 M ruthenium red was without effect (Fig. 3B) and because less Ca 2ϩ was released from filled stores (Fig. 3A), while ryanodine receptors are stimulated by luminal Ca 2ϩ (30). Ca 2ϩ was not released via Ca 2ϩ pumps, because the presence or absence of 2 M thapsigargin, a blocker of pump-mediated Ca 2ϩ release (31), did not affect the hypotonically induced Ca 2ϩ release. Phospholipase A 2 and C blockers (10 M 4-bromophenacyl bromide, 10 M manoalide) had no effect (data not shown), indicating that endogenous production of arachidonic acid or InsP 3 was not involved. Also, modulators of microfilaments or microtubules (10 M phalloidin, 50 M cytochalasin B, 50 M taxol, 10 M demecolcine) had no effect (data not shown).
Divalent cations (2 mM) inhibited the hypotonically induced Ca 2ϩ release with the following order of potency: Ni 2ϩ ϭ Co 2ϩ Ͼ Mn 2ϩ Ͼ Mg 2ϩ Ͼ Ba 2ϩ (Fig. 4A). The inhibition by 2 mM Mg 2ϩ (Fig. 4B) and the other cations (data not shown) was more effective at moderate decreases in tonicity. These ions also decreased the passive InsP 3 -independent Ca 2ϩ leak with the same order of potency (Fig. 4A, inset). We therefore propose that Ca 2ϩ was released through this passive leak pathway. The overall inhibition by these cations was relatively small and actually became even smaller when the [Ni 2ϩ ] was increased from 2 to 10 mM (Fig. 4B, inset).
The hypotonically induced Ca 2ϩ release also occurred in intact A7r5 cells incubated in Ca 2ϩ -free medium. A 40% reduction in extracellular osmolality induced a transient increase in intracellular [Ca 2ϩ ] in 60% of the cells investigated (closed symbols in Fig. 5A). This Ca 2ϩ did not come from outside since the external medium contained no Ca 2ϩ . To discriminate whether this [Ca 2ϩ ] increase represented a Ca 2ϩ release from internal stores or an inhibited Ca 2ϩ extrusion, we have investigated the effect of hypotonic stress on the rate of 45 Ca 2ϩ release from intact A7r5 cells. Fig. 5B shows an enhanced rate of 45 Ca 2ϩ extrusion during the hypotonic shock, indicating that Ca 2ϩ release from intracellular stores and not inhibition of the extrusion caused the [Ca 2ϩ ] increase in the intact cell. 40% of the cells showed no rise in intracellular [Ca 2ϩ ] in response to the hypotonic challenge (open circles in Fig. 5A), although a subsequent vasopressin stimulation (10 M) could release internal Ca 2ϩ .
We conclude that hypotonically induced Ca 2ϩ release through the passive leak pathway is an inherent property of the intracellular stores. Although this phenomenon is not mediated by classical Ca 2ϩ channels such as InsP 3 receptors and ryanodine receptors, it is possible that, in the intact cell, released Ca 2ϩ subsequently activates phospholipase C or A 2 , thereby generating InsP 3 or arachidonic acid and its metabolites. These second messengers may provide a positive feedback loop for the internal Ca 2ϩ release or alternatively activate the necessary mechanisms for volume recovery. Hypotonically Induced Calcium Release 4603